Changeset 8789

Timestamp:

2004-12-13T15:21:50+13:00 (19 years ago)

Author:

mdewsnip

Message:

Better documentation of the extract keyphrases (Kea) code, thanks to Olena.

Location:

trunk/gsdl/perllib

Files:

• plugins/BasPlug.pm (modified) (3 diffs)
• strings.rb (modified) (1 diff)

trunk/gsdl/perllib/plugins/BasPlug.pm

@@ -97 +97 @@
     'type' => "flag",
     'reqd' => "no" },
+      { 'name' => "extract_keyphrases",
+    'desc' => "{BasPlug.extract_keyphrases}",
+    'type' => "flag",
+    'reqd' => "no" },  
+      { 'name' => "extract_keyphrase_options",
+    'desc' => "{BasPlug.extract_keyphrase_options}",
+    'type' => "string",
+    'deft' => "",
+    'reqd' => "no" },  
       { 'name' => "first",
     'desc' => "{BasPlug.first}",
@@ -308 +317 @@
              q^extract_language^, \$self->{'extract_language'},
              q^extract_acronyms^, \$self->{'extract_acronyms'}, 
-             q^extract_keyphrases^, \$self->{'kea'}, #with extra options (UNDOCUMENTED)
-             q^extract_keyphrase_options/.*/^, \$self->{'kea_options'}, #no extra options (UNDOCUMENTED)
+             q^extract_keyphrases^, \$self->{'kea'}, 
+             q^extract_keyphrase_options/.*/^, \$self->{'kea_options'},
              qq^input_encoding/$enc/auto^, \$self->{'input_encoding'},
              qq^default_encoding/$denc/utf8^, \$self->{'default_encoding'},
@@ -875 +884 @@
         $thissection = $doc_obj->get_next_section ($thissection);
     } 
-       
-    if($self->{'kea_options'}) { #if kea options flag is set, call Kea with specified options 
+        
+    #if kea options flag is set, call Kea with specified options 
+    if($self->{'kea_options'}) {
         @list = &Kea::extract_KeyPhrases ($text, $self->{'kea_options'}); 
-    } else { #otherwise call Kea with no options
+    }
+    #otherwise call Kea with no options
+    else { 
         @list = &Kea::extract_KeyPhrases ($text); 
     }

trunk/gsdl/perllib/strings.rb

@@ -522 +522 @@
 BasPlug.extracting_acronyms:extracting acronyms
 
+BasPlug.extract_keyphrases:Extract keyphrases automatically with Kea (default settings).
+
+BasPlug.extract_keyphrase_options:Options for keyphrase extraction with Kea. For example: L2 - length of extracted keyphrases is 2 terms, N5 - 5 keyphrases to extract.
+
 BasPlug.extracting_emails:extracting e-mail addresses